Embodiments of the present specification provide a target recognition method, apparatus, and system. The method comprises: obtaining an image recognition result and a radio frequency recognition result of target objects in a target region, and then determining the distribution of the target objects in the target region according to the radio frequency recognition result and the image recognition result. Since the radio frequency recognition result and the image recognition result are fused, the target objects in the target region can be accurately recognized, so as to improve the recognition accuracy.
Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
1. A method for recognizing a target object, comprising: obtaining an image of a target region; determining an image recognition result of target objects in the target region according to the image; obtaining a radio frequency recognition result of the target objects in the target region; and determining a distribution of the target objects in the target region according to the radio frequency recognition result and the image recognition result; wherein the target region comprises at least one sub-region, and the image recognition result comprises N candidate categories having top-N confidence values of the target objects in each of the at least one sub-region and a number of the target objects in each of the at least one sub-region, and determining the distribution of the target objects in the target region according to the radio frequency recognition result and the image recognition result comprises: respectively matching the N candidate categories and the number of the target objects in each of the at least one sub-region in the image recognition result with categories and a number of the target objects of each category in the radio frequency recognition result, according to a descending order of the top-N confidence values and a descending order of the number of the target objects in each of the at least one sub-region in the image recognition result; and obtaining a category and the number of the target objects in each of the at least one sub-region.
Object recognition and localization. This invention addresses the challenge of accurately identifying and determining the spatial arrangement of target objects within a defined region. The process begins by capturing an image of the target region. An initial image recognition step is performed on this image to identify potential target objects and assign them to candidate categories, along with a confidence score for each category. This image recognition also determines the number of target objects present within the region. Concurrently, a radio frequency recognition process is conducted to obtain information about the target objects, including their categories and the count of objects within each category. The core of the invention lies in fusing these two recognition results. Specifically, the system matches the candidate categories and object counts from the image recognition with the categories and object counts from the radio frequency recognition. This matching is guided by the confidence values from the image recognition and the object counts, prioritizing higher confidence and larger counts. The outcome of this fusion is a refined determination of the category and the precise number of target objects within each sub-region of the target area, thereby establishing their distribution.
2. The method according to claim 1 , wherein determining the image recognition result of the target objects in the target region according to the image comprises: processing the image via a neural network to output the image recognition result of the target objects in the target region.
This invention relates to image recognition systems that identify target objects within a specified region of an image. The problem addressed is the need for accurate and efficient object detection in images, particularly when processing large datasets or real-time applications where computational efficiency is critical. The solution involves using a neural network to analyze an image and determine the recognition results for target objects within a predefined target region. The neural network processes the image data to output the recognition results, which may include object classification, localization, or other relevant information. The method ensures that the recognition process is automated and leverages machine learning techniques to improve accuracy and adaptability across different types of images and objects. The neural network may be pre-trained on a dataset of labeled images to enhance its performance in identifying the target objects. This approach is particularly useful in applications such as surveillance, autonomous navigation, and quality control, where precise and rapid object detection is essential. The system can be integrated into various devices, including cameras, drones, and industrial inspection systems, to provide real-time or batch processing of image data for object recognition tasks. The use of a neural network allows the system to handle complex visual patterns and variations, making it robust against environmental changes and occlusions.
3. The method according to claim 1 , wherein obtaining the radio frequency recognition result of the target objects in the target region comprises: obtaining the radio frequency recognition result of the target objects in the target region via a radio frequency detection antenna arranged in the target region.
This invention relates to radio frequency (RF) recognition systems for identifying and tracking target objects within a defined region. The problem addressed is the need for accurate and efficient detection of objects using RF signals, particularly in environments where visual or other sensing methods may be unreliable or impractical. The method involves obtaining an RF recognition result for target objects within a specified target region. This is achieved by deploying a radio frequency detection antenna within the target region. The antenna captures RF signals emitted or reflected by the target objects, enabling their identification and tracking. The system may also include preprocessing steps to enhance signal quality, such as filtering or amplifying the RF signals before analysis. Additionally, the method may involve comparing the captured RF signals against a database of known signatures to determine the identity or characteristics of the detected objects. The system can be used in applications like inventory management, security monitoring, or automated logistics, where real-time object recognition is essential. The use of RF detection provides advantages over optical or other sensing methods, such as improved performance in low-visibility conditions or when objects are obscured.
4. The method according to claim 3 , wherein the radio frequency recognition result is obtained by reading a radio frequency recognition tag disposed on each of the target objects via the radio frequency detection antenna.
This invention relates to a system for identifying and tracking target objects using radio frequency recognition (RF) technology. The system addresses the challenge of accurately detecting and monitoring objects in environments where visual identification is difficult or impractical, such as in automated warehouses, logistics centers, or manufacturing lines. The method involves using a radio frequency detection antenna to read RF tags attached to each target object. These tags emit or reflect radio frequency signals that are captured by the antenna, allowing the system to identify and locate the objects. The RF tags may be passive, active, or semi-passive, depending on the application requirements. The system processes the signals to determine the presence, position, or movement of the objects, enabling real-time tracking and inventory management. This approach improves efficiency by reducing manual intervention and minimizing errors associated with traditional identification methods. The invention is particularly useful in automated systems where rapid and reliable object recognition is essential. The method ensures accurate detection even in cluttered or obscured environments, enhancing operational reliability and performance.
5. The method according to claim 1 , wherein respectively matching the N candidate categories and the number of the target objects in each of the at least one sub-region in the image recognition result with the categories and the number of the target objects of each category in the radio frequency recognition result according to the descending order of the top-N confidence values and the descending order of the number of the target objects in each of the at least one sub-region in the image recognition result comprises: matching a k-th candidate category of the target objects in an m-th sub-region and a number of the target objects in the m-th sub-region with the categories and the number of the target objects of each category in the radio frequency recognition result; wherein the k-th candidate category is one of the N candidate categories corresponding to a confidence number k; wherein confidence numbers are obtained by ranking the N candidate categories in the m-th sub-region in the image recognition result according to the descending order of the top-N confidence values; wherein the m-th sub-region is a sub-region corresponding to a sub-region number m in the at least one sub-region; and wherein sub-region numbers are obtained by ranking the number of the target objects in each of the at least one sub-region in the image recognition result in a descending order; in response to that the categories in the radio frequency recognition result comprise the k-th candidate category, and the number of the target objects of the k-th candidate category in the radio frequency recognition result is greater than zero, determining that a category of the target objects in the m-th sub-region is the k-th candidate category of the target objects in the m-th sub-region in the image recognition result and the number of the target objects in the m-th sub-region is the number of the target objects in the m-th sub-region in the image recognition result; and deducting the number of the target objects in the m-th sub-region from the number of the target objects of the k-th candidate category in the radio frequency recognition result to obtain an adjusted radio frequency recognition result.
This invention relates to a method for improving object recognition accuracy by combining image recognition and radio frequency (RF) recognition results. The problem addressed is the potential inaccuracy in standalone image or RF recognition systems, particularly in scenarios where objects may be occluded or misclassified. The solution involves cross-referencing and refining recognition results from both modalities to enhance reliability. The method processes an image recognition result that includes candidate categories for target objects, each with associated confidence values, and counts of objects in sub-regions of the image. It also processes an RF recognition result that includes categories of detected objects and their counts. The method matches the top-N candidate categories from the image recognition result with the RF recognition result, prioritized by descending confidence values and object counts per sub-region. For each sub-region, the method checks if the RF recognition result contains the k-th candidate category (ranked by confidence) and if the object count for that category in RF is non-zero. If both conditions are met, the method assigns the k-th candidate category to the sub-region and updates the RF recognition result by deducting the matched object count. This iterative process refines the recognition results, ensuring consistency between the two modalities. The adjusted RF recognition result is then used for further analysis or decision-making. This approach improves accuracy by leveraging the strengths of both image and RF recognition systems.
6. The method according to claim 5 , further comprising: in response to that the categories in the radio frequency recognition result do not comprise the k-th candidate category or in response to that the categories in the radio frequency recognition result comprise the k-th candidate category but the number of the target objects of the k-th candidate category in the radio frequency recognition result is not greater than zero, determining that matching is unsuccessful; and determining the m-th sub-region as a sub-region to be matched.
This invention relates to radio frequency recognition systems, specifically improving the accuracy of object categorization and matching in such systems. The problem addressed is the unreliable identification of target objects when radio frequency recognition results do not clearly match predefined candidate categories or when the count of detected objects in a category is insufficient. The method involves analyzing radio frequency recognition results to determine if a target object belongs to a specific candidate category. If the recognition results do not include the expected category or if the detected number of objects in that category is zero, the system declares the matching unsuccessful. In response, the system selects a different sub-region for further matching, ensuring that the recognition process continues even when initial attempts fail. This approach enhances the robustness of object detection by dynamically adjusting the search area based on recognition outcomes, reducing false negatives and improving overall system reliability. The method is particularly useful in environments where radio frequency signals may be noisy or where objects are partially obscured, ensuring consistent performance in challenging conditions.
7. The method according to claim 6 , wherein when determining the m-th sub-region as the sub-region to be matched, the method further comprises: in response to that the number of the target objects of each category in the radio frequency recognition result is not greater than zero, determining that one or more candidate categories, in the N candidate categories in the sub-region to be matched, appear in the categories of the radio frequency recognition result, taking the candidate category having a maximum confidence value in one or more appeared candidate categories as the category of the target objects in the sub-region to be matched, and taking the number of the target objects in the sub-region to be matched in the image recognition result as the number of the target objects in the sub-region to be matched.
This invention relates to object recognition systems that combine radio frequency (RF) and image recognition technologies to improve accuracy in identifying and counting target objects. The problem addressed is the discrepancy between RF and image recognition results, particularly when RF detection fails to identify certain objects, leading to incomplete or inaccurate data. The method involves dividing an area into multiple sub-regions and performing both RF and image recognition within each sub-regions. When RF recognition fails to detect any target objects (i.e., the count is zero for all categories), the system checks the candidate categories in the sub-region against the categories identified by image recognition. If any candidate categories from the sub-region match those in the image recognition result, the system selects the candidate category with the highest confidence value as the category for the target objects in that sub-region. The count of target objects from the image recognition result is then used as the final count for that sub-region. This approach ensures that even when RF recognition fails, the system can still rely on image recognition to provide accurate category and count information, improving overall system reliability. The method is particularly useful in applications where objects may be obscured or RF signals are weak, such as in inventory management or automated sorting systems.
8. The method according to claim 6 , wherein when determining the m-th sub-region as the sub-region to be matched, the method further comprises: in response to that the number of the target objects of each category in the radio frequency recognition result is not greater than zero, determining that none of the N candidate categories in the sub-region to be matched appears in the categories of the radio frequency recognition result, taking the candidate category having a maximum confidence value as the category of the target objects in the sub-region to be matched, and taking the number of the target objects in the sub-region to be matched in the image recognition result as the number of the target objects in the sub-region to be matched.
This invention relates to a method for improving object recognition accuracy in systems that combine radio frequency (RF) and image recognition technologies. The problem addressed is the discrepancy between RF recognition results, which may not detect all target objects, and image recognition results, which may misclassify objects. The method aims to resolve inconsistencies by refining the categorization and counting of target objects in sub-regions of a monitored area. The method involves dividing the monitored area into multiple sub-regions and analyzing each sub-region individually. For each sub-region, the method compares the RF recognition results with the image recognition results. If the RF recognition results indicate that no target objects of any category are present in a sub-region, the method selects the candidate category with the highest confidence value from the image recognition results as the final category for that sub-region. Additionally, the method uses the object count from the image recognition results for that sub-region, overriding the RF recognition results. This approach ensures that even when RF recognition fails to detect objects, the system still provides accurate categorization and counting based on image recognition data, improving overall reliability. The method is particularly useful in applications requiring precise object tracking, such as inventory management or security monitoring.
9. The method according to claim 6 , wherein when determining the m-th sub-region as the sub-region to be matched, the method further comprises: in response to that the number of the target objects of at least one category in the radio frequency recognition result is greater than zero, taking the category with a first maximum number of the target objects in the radio frequency recognition result as the category of the target objects in the sub-region to be matched with a second maximum number of the target objects in the image recognition result, and taking the number of the target objects in the sub-region to be matched with the second maximum number of the target objects in the image recognition result as the number of the target objects in the sub-region to be matched with the second maximum number of the target objects in the image recognition result; and deducting the number of the target objects in the sub-region to be matched with the second maximum number of the target objects in the image recognition result from the number of the target objects in the category with the first maximum number of the target objects in the radio frequency recognition result to obtain an adjusted radio frequency recognition result.
This invention relates to a method for improving object recognition accuracy by combining radio frequency (RF) and image recognition results. The method addresses the challenge of discrepancies between RF-based and visual-based object detection, particularly in scenarios where objects may be occluded or misclassified by one or both systems. The method involves analyzing a scene divided into multiple sub-regions. For each sub-region, the method compares the number of detected objects from RF recognition with those from image recognition. If at least one category of objects is detected in the RF recognition result, the category with the highest count (first maximum) in the RF result is matched with the category having the highest count (second maximum) in the image recognition result. The number of objects in the matched sub-region from the image recognition is then used to adjust the RF recognition result by deducting this number from the RF count. This adjustment refines the RF recognition result, improving accuracy by resolving mismatches between the two detection methods. The method ensures that the combined recognition result is more reliable by dynamically aligning the counts of objects detected by RF and image recognition, particularly in cases where one system may have missed or misidentified objects. This approach enhances object tracking and identification in environments where both RF and visual data are available.
10. A system for recognizing a target object, comprising: an image collection apparatus, configured to obtain an image of a target region; a radio frequency detection apparatus, configured to obtain a radio frequency recognition result of target objects in the target region; and a processor, configured to determine an image recognition result of the target objects in the target region according to the image, and determine a distribution of the target objects in the target region according to the radio frequency recognition result and the image recognition result; wherein the target region comprises at least one sub-region, and the image recognition result comprises N candidate categories having top-N confidence values of the target objects in each of the at least one sub-region and a number of the target objects in each of the at least one sub-region; and determining the distribution of the target objects in the target region according to the radio frequency recognition result and the image recognition result comprises: respectively matching the N candidate categories and the number of the target objects in each of the at least one sub-region in the image recognition result with categories and a number of the target objects of each category in the radio frequency recognition result according to a descending order of the top-N confidence values and a descending order of the number of the target objects in each of the at least one sub-region in the image recognition result; and obtaining a category and the number of the target objects in each of the at least one sub-region.
The system is designed for recognizing and distributing target objects within a defined region using both image and radio frequency (RF) detection. The system addresses the challenge of accurately identifying and localizing objects in environments where visual data alone may be insufficient, such as in low-light or occluded conditions. The system includes an image collection apparatus to capture images of the target region and an RF detection apparatus to obtain RF recognition results of objects within that region. A processor analyzes the image to generate an image recognition result, which includes the top-N most likely categories (N candidate categories) of objects in each sub-region of the target region, along with the count of objects in each sub-region. The processor also uses the RF recognition result, which provides categories and counts of objects detected via RF signals. To determine the distribution of objects, the system matches the N candidate categories and object counts from the image recognition result with the categories and counts from the RF recognition result. This matching is performed in descending order of confidence values from the image recognition and descending order of object counts. The result is a refined distribution of object categories and their quantities in each sub-region, combining the strengths of both visual and RF-based detection methods. This approach improves accuracy in object recognition and localization in complex environments.
11. The system according to claim 10 , wherein respectively matching, by the processor, the N candidate categories and the number of the target objects in each of the at least one sub-region in the image recognition result with the categories and the number of the target objects of each category in the radio frequency recognition result according to the descending order of the top-N confidence values and the descending order of the number of the target objects in each of the at least one sub-region in the image recognition result comprises: matching a k-th candidate category of the target objects in an m-th sub-region and a number of the target objects in the m-th sub-region with the categories and the number of the target objects of each category in the radio frequency recognition result; wherein the k-th candidate category is one of the N candidate categories corresponding to a confidence number k; wherein confidence numbers are obtained by ranking the N candidate categories in the m-th sub-region in the image recognition result according to the descending order of the top-N confidence values; wherein the m-th sub-region is a sub-region corresponding to a sub-region number m in the at least one sub-region; and wherein sub-region numbers are obtained by ranking the number of the target objects in each of the at least one sub-region in the image recognition result in a descending order; in response to that the categories in the radio frequency recognition result comprise the k-th candidate category, and the number of the target objects of the k-th candidate category in the radio frequency recognition result is greater than zero, determining that a category of the target objects in the m-th sub-region is the k-th candidate category of the target objects in the m-th sub-region in the image recognition result and the number of the target objects in the m-th sub-region is the number of the target objects in the m-th sub-region in the image recognition result; and deducting the number of the target objects in the m-th sub-region from the number of the target objects of the k-th candidate category in the radio frequency recognition result to obtain an adjusted radio frequency recognition result.
This invention relates to a system for improving object recognition accuracy by combining image recognition and radio frequency (RF) recognition results. The system addresses the challenge of accurately identifying and counting objects in an environment where visual data alone may be insufficient or ambiguous. The system processes an image recognition result that includes candidate categories for target objects and their respective confidence values, along with the number of objects detected in sub-regions of the image. It also processes an RF recognition result that provides categories of detected objects and their counts. The system matches the top-N candidate categories from the image recognition result with the RF recognition result, prioritized by descending confidence values and object counts in each sub-region. For each sub-region, the system checks if the RF recognition result includes the k-th candidate category (ranked by confidence) and if the count of that category in the RF result is positive. If so, it confirms the category and count from the image recognition result for that sub-region and updates the RF result by deducting the confirmed count. This iterative process refines the recognition results, enhancing accuracy by cross-referencing visual and RF data. The system is particularly useful in applications requiring precise object identification and counting, such as inventory management or surveillance.
12. The system according to claim 11 , wherein the processor is further configured to: in response to that the categories in the radio frequency recognition result do not comprise the k-th candidate category or in response to that the categories in the radio frequency recognition result comprise the k-th candidate category but the number of the target objects of the k-th candidate category in the radio frequency recognition result is not greater than zero, determine that matching is unsuccessful; and determine the m-th sub-region as a sub-region to be matched.
This invention relates to a system for radio frequency (RF) recognition and object categorization, addressing challenges in accurately identifying and classifying target objects in a monitored area. The system uses RF signals to detect and categorize objects, but may encounter difficulties when the recognition results do not match expected categories or when the count of detected objects in a specific category is insufficient. To resolve this, the system includes a processor that evaluates the RF recognition results against predefined candidate categories. If the RF recognition results do not include a particular candidate category (the k-th category) or if the detected count of objects in that category is zero or negative, the system determines that the matching process has failed. In such cases, the system selects an alternative sub-region (the m-th sub-region) for further matching attempts. This approach ensures robust object recognition by dynamically adjusting the matching process when initial results are inconclusive or incomplete. The system may also include additional components, such as an RF signal transmitter and receiver, to generate and analyze the RF signals used for object detection and categorization. The processor further processes the RF recognition results to refine object identification and classification, improving accuracy in dynamic environments.
13. The system according to claim 12 , wherein when determining the m-th sub-region as the sub-region to be matched, the processor is further configured to: in response to that the number of the target objects of each category in the radio frequency recognition result is not greater than zero, determine that one or more candidate categories, in the N candidate categories in the sub-region to be matched, appear in the categories of the radio frequency recognition result, take the candidate category having a maximum confidence value in one or more appeared candidate categories as the category of the target objects in the sub-region to be matched, and take the number of the target objects in the sub-region to be matched in the image recognition result as the number of the target objects in the sub-region to be matched.
This invention relates to a system for matching target objects detected in an image with categories identified through radio frequency (RF) recognition. The system addresses the challenge of accurately associating objects in an image with their corresponding categories when RF recognition results are incomplete or ambiguous. The system divides an image into multiple sub-regions and processes each sub-region to determine the most likely category for any detected objects. When RF recognition fails to detect any objects of a particular category in a sub-region, the system evaluates candidate categories from the RF recognition results. It selects the candidate category with the highest confidence value from those that appear in the RF recognition results and assigns it to the sub-region. The system then uses the number of objects detected in the image recognition result for that sub-region as the count for the target objects. This approach ensures that even when RF recognition is incomplete, the system can still accurately categorize and count objects in the image by leveraging both RF and image recognition data. The system improves object recognition accuracy in scenarios where RF signals may be weak or obstructed, enhancing reliability in applications such as inventory management, security monitoring, or automated sorting.
14. The system according to claim 12 , wherein when determining the m-th sub-region as the sub-region to be matched, the processor is further configured to: in response to that the number of the target objects of each category in the radio frequency recognition result is not greater than zero, determine that none of the N candidate categories in the sub-region to be matched appears in the categories of the radio frequency recognition result, take the candidate category a maximum confidence value as the category of the target objects in the sub-region to be matched, and take the number of the target objects in the sub-region to be matched in the image recognition result as the number of the target objects in the sub-region to be matched.
This invention relates to a system for matching target objects detected in an image with categories identified through radio frequency (RF) recognition, particularly when no RF-detected objects are present in a sub-region of the image. The system addresses the challenge of accurately categorizing and counting objects in scenarios where RF recognition fails to detect any targets, ensuring reliable object tracking and classification. The system processes an image divided into multiple sub-regions, each containing candidate categories of target objects. When RF recognition does not detect any objects (i.e., the count of target objects for each category is zero) in a specific sub-region, the system defaults to image recognition results. It selects the candidate category with the highest confidence value from the image recognition output as the category for that sub-region. Additionally, the system uses the object count from the image recognition result for the sub-region, overriding the RF recognition data. This approach ensures continuity in object tracking when RF recognition is unreliable or unavailable, improving accuracy in mixed-sensor environments. The system integrates RF and image recognition to enhance object detection robustness, particularly in dynamic or RF-challenged scenarios.
15. The system according to claim 12 , wherein when determining the m-th sub-region as the sub-region to be matched exists, the processor is further configured to: in response to that the number of the target objects of at least one category in the radio frequency recognition result is greater than zero, take the category with a first maximum number of the target objects in the radio frequency recognition result as the category of the target objects in the sub-region to be matched with a second maximum number of the target objects in the image recognition result, and take the number of the target objects in the sub-region to be matched with the second maximum number of the target objects in the image recognition result as the number of the target objects in the sub-region to be matched with the second maximum number of the target objects in the image recognition result; and deduct the number of the target objects in the sub-region to be matched with the second maximum number of the target objects in the image recognition result from the number of the target objects in the category with the first maximum number of the target objects in the radio frequency recognition result to obtain an adjusted radio frequency recognition result.
This invention relates to a system for matching and adjusting target object counts between radio frequency (RF) recognition and image recognition results in a monitored area. The system addresses the challenge of discrepancies between RF and image-based object detection, particularly in scenarios where objects may be occluded or misclassified. The system divides the monitored area into sub-regions and compares the number of detected objects in each sub-region between RF and image recognition systems. When a sub-region is identified for matching, the system determines the category of objects with the highest count in the RF recognition result and compares it to the category with the highest count in the image recognition result for the same sub-region. If the RF recognition result indicates at least one object of a given category, the system adjusts the RF count by deducting the number of objects matched in the image recognition result, ensuring consistency between the two detection methods. This adjustment helps improve accuracy in object tracking and classification by resolving discrepancies between RF and visual data. The system is particularly useful in applications like inventory management, security monitoring, or automated surveillance where reliable object detection is critical.
16. A non-transitory computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the processor is caused to perform operations comprising: obtaining an image of a target region; determining an image recognition result of target objects in the target region according to the image; obtaining a radio frequency recognition result of the target objects in the target region; and determining a distribution of the target objects in the target region according to the radio frequency recognition result and the image recognition result; wherein the target region comprises at least one sub-region, and the image recognition result comprises N candidate categories having top-N confidence values of the target objects in each of the at least one sub-region and a number of the target objects in each of the at least one sub-region; and determining the distribution of the target objects in the target region according to the radio frequency recognition result and the image recognition result comprises: respectively matching the N candidate categories and the number of the target objects in each of the at least one sub-region in the image recognition result with categories and a number of the target objects of each category in the radio frequency recognition result according to a descending order of the top-N confidence values and a descending order of the number of the target objects in each of the at least one sub-region in the image recognition result; and obtaining a category and the number of the target objects in each of the at least one sub-region.
This invention relates to a system for accurately identifying and distributing target objects within a monitored region using both image recognition and radio frequency (RF) recognition techniques. The system addresses the challenge of improving object detection accuracy by combining visual and RF data, which can be more reliable than either method alone. The system captures an image of a target region, which may be divided into sub-regions, and processes the image to identify target objects, generating a list of candidate categories with associated confidence values and counts for each sub-region. Simultaneously, the system obtains an RF recognition result, which provides categorical and numerical data for the target objects. The system then matches the image recognition results with the RF recognition results by prioritizing the highest-confidence categories and the highest object counts from the image data. This matching process refines the distribution of objects across the sub-regions, ensuring accurate identification and localization. The combined approach enhances detection reliability, particularly in environments where visual or RF data alone may be insufficient. The system is implemented via a computer program stored on a non-transitory storage medium, executed by a processor to perform the described operations.
Cooperative Patent Classification codes for this invention. Click any code to explore related patents in that topic.
April 18, 2020
April 5, 2022
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.